JPH04165425A - Method and device for extracting nesting relation in symbol string - Google Patents

Abstract

PURPOSE:To easily extract nesting relation information on the depth of nesting, the appearance order of the nesting, branch condition order, etc., from symbol string in nesting structure by processing the symbol strings sequentially. CONSTITUTION:Input symbol strings in line units which are stored in a symbol storage part 2 are sent to a key word extraction part 3, which extracts a key word defining the attribute of the nesting. A nesting state detection part 4 detects a symbol string indicating the start or end of the nesting and updates the nesting depth D stored in a nesting depth storage part 6 and the appearance order and condition branch order of the nesting at the depth D which are stored in a nesting order depth storage part 7. A nesting relation output part 8 adds the appearance order and conditional branch order of the nesting at the nesting depth D to each of the input symbol strings in line units and outputs them. Consequently, not only depth information on the nesting, but also information on mutual order relation information between nesting parts can be extracted and outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method and apparatus for extracting nested relationships from a symbol string written in a programming language such as C language or LISP language.

[Prior Art] A symbol string written in a programming language such as C language or LISP language is called a program. If a block of lines, such as a group of lines to be executed specified by a repeat directive in a program or a group of lines to be executed for each condition in a conditional judgment statement, is an area surrounded by nesting, the depth of nesting and mutual relationships can be controlled. Understanding this is very useful in understanding the flow of a program.

Conventionally, methods for understanding the structure of a program include simply displaying the nesting depth, or repeatedly using graphical information such as HCP charts and PAD, and assigning specific symbols to conditional branches, etc. There is a way to display the flow.

[Problem to be Solved by the Invention] An example of a method for displaying only the depth of nesting is shown in the case of LISP. Since '(' is the start symbol of nesting, and ')'' is the end of nesting, When ``('' appears, add 1 to the nesting depth and ``)'.

The nesting depth can be easily displayed by subtracting the nesting depth by 1 each time . However, since the only information displayed is the depth of nesting, the interrelationships between lines are not displayed, and if the nesting relationships become somewhat complex, it becomes difficult to understand the flow of the entire program.

On the other hand, in the case of the method of displaying the flow of a program using diagrams, all the details of the program are written in diagrams, so the structure of the program is not lost, unlike when only the nesting depth is displayed. However, since everything is represented in diagrams, the amount of information written per sheet decreases, and there is a drawback that the flow of the program cannot be understood from a broader perspective.

An object of the present invention is to extract the nesting relationship at any position in a symbol string based on symbols that start and end nesting designated in advance, and to determine the depth of nesting and the order with other nests. The object of the present invention is to provide a method and device for easily expressing relationships.

[Means for Solving the Problem] In order to achieve the above object, in claim (1), a keyword that defines a nested attribute and a symbol string that starts/ends nesting are selected from symbol strings. In the method of extracting the nesting relationship at an arbitrary position, the first process is to extract the nesting depth D at an arbitrary position of a symbol string, and the keyword defining the nesting attribute is extracted, The appearance order at the depth D includes at least a second process of counting S o and a third process of counting the order CSD of conditional branches within the nest at the depth D, In claim (2), the state is extracted from a symbol string at an arbitrary position based on a keyword that defines a nested attribute and a symbol string that starts/ends nesting. A device for extracting a nested relationship in , an input section for inputting a symbol string from the outside, and an input symbol string storage section for storing the symbol string input from the input section on a line-by-line basis with an arbitrary length as a line. , a keyword storage unit that stores keywords, and a nesting depth storage unit that stores nesting depth D.
Appearance order KS of keywords at the nesting depth D
, a nested order storage unit that stores the order CSD of conditional branches within the nest; and a keyword is extracted from the input symbol string stored in the input symbol string storage unit, and the keyword is stored in the keyword storage unit. a keyword extraction unit to
A symbol string indicating the start/end of nesting is detected from the input symbol string, and based on the contents of the nesting depth storage section and the nesting order storage section, the nesting depth D and the appearance of the keyword at the depth D are detected. a nested state detection unit that calculates the order KS o and the order C8o of conditional branches within the nest and stores them in the nesting depth storage unit and the insertion order storage unit respectively; The nesting depth D stored in the nesting depth storage unit, the appearance order KSD and the conditional branch order CSD stored in the nesting order storage unit are added to the symbol string in row units and output. and the nested relationship output section.

[Function] In the present invention, a specific symbol is used as a line delimiter Cr, and C
If the symbol string between r is one line, the number of lines in one line is Ln (
In a program composed of m lines with 1≦Ln), a group of symbol strings NSi (
1≦1) and the symbol string group NEj (1
≦j), not only nesting depth information but also mutual order relationship information between nests is extracted and output. Note that basically, even without the line delimiter Cr, it is possible to perform processing with any symbol string length, so the concept of a line is not essential.

[Example〕 An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows the processing procedure of an embodiment of the method of the present invention. In Figure 1, 'depth' is a variable name, '
The "order of appearance [depth]" indicates the "depth"-th value of the matrix variable with the name "order of appearance".The variable name used here is arbitrary, so even if the variable name is changed, , the essence remains the same.Furthermore, to make the explanation easier, we will show the case where processing is performed line by line, but since it is possible to process with any symbol string length, the concept of line is essential. isn't it.

(1) First, set the nesting "depth" to ○ and the appearance order [1] to 0 (step 101).

(ii) Read one line (step 102), check whether the reading is completed (step 103), and if it is completed, the process ends.

(iii) If the reading is not completed, check whether the keyword exists in one line (step 104), and if it exists, store the keyword (step 10).
5).

(iv) Check whether nesting has started (step 106). At the start, increase "depth" by 1 (
step 107), the appearance order [depth is increased by one (
Step 108), condition order [depth] is set to 1 (step] 09), and appearance order [depth + 1] is set to 0 (step 110). After that, return to (jl).

(V) If it is not the start of a nest, check whether it is the start of a conditional clause (step 1.11); at the start, increase the conditional order [depth by 1 (step 112); (ii)
Return to

(vi) If it is not the start of a conditional clause, check whether it is the end of nesting (step 113). Then, at the end of nesting, decrease the ``depth'' by 1 (step 1
14), return to (11).

Next, a specific example will be explained using FIG. 2. The keywords are "if (...)" and "for (...)". If ``('' appears alone immediately after a keyword, it is considered the start of nesting.

”) If Pa appears alone, it is the end of nesting.“
)else <" is the start of a conditional clause. Processing that transfers control from one nest to another is not allowed.

As preprocessing, the depth ′ and the first term of the appearance order variable (appearance order [1 item)] are set to O.

(1) Keyword ""if(...)" in the -th line
is detected, and the keyword ''if' is stored in key [1].

(11) In the second nature, ' is added and stored as 1, 1 is stored in the condition order [1], and O is stored in the appearance order [2].

(iii) In the third line, ")effise(" is detected, and as condition part processing, condition order [+, content 1 of E
Add 1 to and store it as 2.

(1v) In the fourth line, the keyword “if (...)
” is detected, the keyword “1f” is stored in key [2], “(” is detected, and as a process at the start of nesting,
Add 1 to the content of ``depth'' and store it as 2, add 1 to the content of 2 and store it as 1, store the content of 2 as 1, store 1 in the condition order [2], and change the appearance order [ Store 0 in 3.

Hereinafter, the processing will proceed according to the processing procedure shown in FIG.

An example of the processing result in the case of FIG. 2 is shown in FIG. In this example, to reduce the amount of information, the order of appearance [depth], k
ey [depth] and condition order [depth] are output as bare values for each depth and added to the program from the left. The essential point is that each line of the program is output along with the order in which that line appears; c. The method of expressing the order of appearance, etc. does not matter.

FIG. 4 shows a block diagram of an embodiment of the apparatus of the present invention.

The symbol string inputted from the input section 1 is sent to the input symbol string storage section 2 row by row via the signal line ll and is stored therein.

The input symbol string stored in the symbol storage section 2 in units of rows is sent to the keyword extraction section 3 via the signal line 12. The keyword extraction unit 3 extracts a keyword that defines a nested attribute, sends the keyword to the keyword storage unit 5 via the signal a14, and simultaneously sends a symbol string to the nested state detection unit 4 via the signal line I3. The nesting depth storage unit 6 stores the nesting depth D in the middle of processing, and similarly, the nesting order storage unit 7 stores S D and the conditional branch order C8l in the order of appearance at the depth. ing.

The nesting state detection section 4 detects a symbol string indicating the start/end of nesting, obtains the nesting depth D from the nesting depth storage section 6 via the signal line ]5, and also obtains the nesting depth D from the nesting depth storage section 6 through the signal line ]5.
After obtaining the nest appearance order KSD and conditional branch order CSD at the depth D via the signal line 16, the depth D is updated and sent to the nest depth storage unit 6 via the signal line 15 (H The appearance order KS11. and the conditional branching order CSD are updated and sent to the nested order storage unit 7 via the signal line 16.The nested opening output unit 8 receives the input symbol string via the signal line 20 and sends it to the nested order storage unit 7 via the signal line 20. The keyword is obtained via the signal line 18, the nesting depth information is obtained via the signal line 18, and the nesting order relationship is obtained via the signal line 19. KSD and conditional branch order C3ll+ are added and output.

The above process is repeated until there are no more symbol strings input from the input unit 1.

〔Effect of the invention〕

As explained above, according to the present invention, by sequentially processing symbol strings, nesting relationships such as nesting depth, nesting appearance order, conditional branching order, etc. can be easily obtained from symbol strings having a nested structure. It becomes possible to extract information.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of a procedure for extracting a nested relationship according to the method of the present invention, FIG. 2 is a diagram showing a specific example of the nested relationship extraction process according to the present invention, and FIG. FIG. 4, which is a diagram showing an example of the output obtained, is a block diagram of an embodiment of the apparatus of the present invention. 1... Keyword, 2... Input symbol string storage section, 3
...Keyword extractor, 4...Nested state detector, 5...Keyword memory, 6...Nesting depth memory, 7...Nested order memory, 8...Nested relationship output unit . Figure 2 1i i dried ()111(...
・ 1121 jin 1se (... 1i21i1 'If (') jin... 1iZ
IiZ 1 elfe (-. ljZ J tit 2i1 '+'r () (--+i
z 2iz l P-+se (,,,
, -1122i21'+f if ()
(,-?i22iZ IiZ) el
se (,,,,. ?i22'+2 3 1iz) 7i23i+ 'I し (
) (...1iZ 3i2 1 els
e i,,,,. 2t1 child or (”l
(,,,. 2 I Tit if
() (middle school 2fl IiZ'
J elSe (-, -pi appearance order 4th ry: f

Claims (2)

[Claims] (1) A method for extracting a nesting relationship at an arbitrary position from a symbol string based on a keyword that defines a nested attribute and a symbol string that starts/ends nesting, a first process of extracting the nesting depth D in the nesting depth D; a second process of extracting a keyword defining the nested attribute and counting the appearance order KS_D at the nesting depth D; A nested relationship extraction method in a symbol string, characterized in that the nested state at an arbitrary symbol position is extracted, the method comprising at least a third process of enumerating the order CS_D of conditional branches within the symbol string. (2) A symbol string is input from the outside in a device that extracts nesting relationships at arbitrary positions from symbol strings based on keywords that define nested attributes and symbol strings that start and end nesting. an input symbol string storage section that stores the symbol string input from the input section line by line with an arbitrary length; a keyword storage section that stores keywords; and a nesting depth D. a nesting depth storage unit that stores the keyword appearance order KS at the nesting depth D;
a nested order storage unit that stores __D and the conditional branch order CS_D within the nest; a keyword is extracted from the input symbol string stored in the input symbol string storage unit, and the keyword is stored in the keyword storage unit; a keyword extraction unit that detects a symbol string indicating the start/end of nesting from the input symbol string, and calculates the nesting depth D and the depth based on the contents of the nesting depth storage unit and the nesting order storage unit; a nested state detection unit that calculates the appearance order KS_D of keywords in the number D and the order CS_D of conditional branches in the nest, and stores them again in the nesting depth storage unit and the nesting order storage unit, respectively; and the input symbol string storage unit The nesting depth D stored in the nesting depth storage unit and the appearance order KS_D stored in the nesting order storage unit for the row-by-line symbol string stored in
and a nested relationship output unit that adds and outputs a conditional branch order CS_D.